The research aims at establishing a relationship between the fear of crime, gender and prior victimization. Prior research indicates that there have been weak links between the latter factors hence the research aims at assorting the three parameters.

The research will be conducted through structural interviews and secondary sources. The respondents will be determined through non probability sampling with strict adherence to gender balance. The Likert scale will be used for the structured interview. Thereafter, the data will be analyzed though bar charts and standard deviation for the two independent variables: gender and prior victimization will be calculated in relation to the dependent variable; fear of crime. The two values of standard deviation will be compared and an inference made.

The research will be managed by a project coordinator with the help of four other team members. Expenses include supplies, traveling expenses and personnel logistics; they all sum up to $ 6, 770.


Research conducted by Skogan (1986) found that the fear of crime creates negative psychological effects in a community. Consequently, stakeholders in the crime prevention sector need to equip themselves with knowledge surrounding this topic. They need to know factors linked to the fear of crime and the magnitude or relative importance of each. However, this essay will only examine two factors; gender and prior victimization. Examining the relationship between fear of crime, gender and prior victimization is quite instrumental in crime prevention strategies. This is because at the end of the research, there will be empirical evidence to help law enforcement officers, government agencies and stakeholders in the crime prevention unit. (Hale 1996)


Morris et al (2003) conducted a survey in New Zealand linking numerous demographic factors to crime victimization. They found that gender has a significant effect on the perceptions of safety hence fear of crime. McGarrell et al (1997) also found that when one had been a victim of crime in the past, chances were that they would depict greater fear of the same in the future.  Hale (1996) adds that some disparities exist among people who had never been victimized before and those who were at greater risk.

Akers et al (1987) tried examining the link between victimization and fear of crime. Their results were not very convincing because the relationship was quite weak. Similarly, Bennet & Flavin (1994) also came up with weak conclusions. However, there were other researchers that managed to come up with links relating the above subjects. For instance, Stiles et al (2003) found a strong link between victimization and crime. These findings were affirmed by Acierno et al (2004) and Moore and Shepherd (2007).  As it can be seen, there are mixed perceptions about the effect of victimization on fear crime. The research intends on clarifying this matter and linking this factor to gender.

In order to establish a relationship between the fear of crime and other factors, it is necessary o understand what kind of indicators ‘denote fear of crime’. This allows better and more proficient measurement. Surveys conducted by the Australian Bureau of Statistics in (2006) and by Statistics New Zealand (2006) employed similar techniques. These surveys considered; how safe people felt when at home, when traveling or when walking alone within specific communities. Conversely, Ferraro and LaGrange (1988) argue that these factors are invalid. They believe that these factors are not directly linked to crime or to fear. For instance, by asking people about their feelings of safety when walking home, researchers tend to include other issues such as the fear of an accident or a heart attack. These factors are ambiguous and lack direct relation to crime. Ferraro and La Grange (1988) also argue that most of the issues indicate quality of life rather than fear of crime. This is because they indicate the kind of risk present in certain communities or how safe particular individuals really are. These individuals base their understanding of safety upon their feelings of uncertainty, concern and anxiousness.

In light of these arguments, some authors came up with better indictors of fear of crime. Warr (1984) asked people about their fear of crime in relation to a number of crimes. Figgie (1983) concurs with this argument; he believes that questions should directly talk about the fear of crime in order to depict accurate results.


The research questions chosen above will require validation of two major factors then it will eventually necessitate linking those factors together. At this point, there will be a need to move from theory to operationalization of the concept. The literature review conducted above has resulted in the creation of interlinked logical assumptions and propositions. For instance, it has been established that there is a link between gender and fear of crime but this link is not clearly understood. Additionally, it has been assumed that prior victimization is related to fear of crime but the extent of this relationship needs to be established. Therefore, it is essential that the report clearly chooses the best indicators of the latter three variables i.e. gender, fear of crime and prior victimization. Consequently, measuring these major concepts is the major challenge hen trying to construct the research questions. (Babbie, 2005)

It should be noted that various methods of asking one question are possible and each approach would result in totally different responses hence different analyses and results. This research questions may be asked in such way that they solicit open ended responses. For instance, the respondents may be asked what their fear of crime is in relation to their community. Then responses will be evaluated on the basis of gender. This is called the nominal approach. On the other hand, one can evaluate level of crime by asking respondents to answer yes or no with if their fear of crime of a high or low degree respectively. This is what is called the ordinal level of setting research questions. (Meyers et al, 2005)

The latter method is more appropriate in situations where the research questions are sociological. Because this is a sociological research then the latter method could be used to vary the research questions. It could allow better manipulation of results because of the simplicity of the responses. Open ended questions may require too much statistical maneuver and this may eventually interfere with the final results. Additionally, this method is also appropriate because the research will be considering gender as a factor. Experts believe that when one is examining another indicator in between, then the ordinal approach is more appropriate. (Babbie, 2005)


The main reasons why structured interviews were chosen for this research are: Participants give very good response rates. Additionally, some interviewees may not understand particular questions but since the interviewer is in direct contact with the respondent, then chances are that they can iron out these anomalies. Besides these, it reduces bias on the part if the interviewer because there are preset questions to follow. Also, because there is uniform assortment of questions, then chances are that the results can be easily compared. Also such an approach is quite useful when quantifying data. (Meyers et al, 2005)

However, there are also some disadvantages to this method of data collection. First of all, it will require more time than simply mailing questionnaires to participants. This deficiency will be minimized by delegating duties and also by planning the design well. Besides that the structured interview format can prove to be quite expensive is administered in large portion or are. Consequently, there is a need to make sure that most of the responses are spread over a huge small size. However, this inadequacy will be reduced by reducing the sample size. Interviewers themselves may have some bias when recording results but this can be ironed out by conducting some form of training prior to the actual interviews. (Likert, 1932)

Likert Scaling is quite appropriate in this scenario because ‘fear of crime’ as a main factor is quite difficult to quantify, therefore using Likert items will help in clarifying out some of the arguments. Likert scales measure negative or positive responses and this will be quite instrumental in the data analysis stage. This is because all the answers given under the ‘disagree’ and ‘strongly disagree’ categories can be summed up to represent negative responses while those ones given under the ‘agree’ and ‘strongly disagree’ categories can also be summed to represent the positive responses. It should be noted that there is room for distortion in this kind of approach; some respondents may shy away form extreme respondents in order to appear socially acceptable. Furthermore, some may tend to give positive responses in order to make their community look safe and free form crime. This disadvantage will be minimized by balancing out the number of negative and positive statements to be addressed in the interviews. At the en of the research, all the Likert questions will be added up in order to come up with a final response related to fear of crime. It should be noted that adding up the statements is only permissible is all the question were asked using the Likert response. ‘Addition’ in this case will not be done literally. It will be achieved through variance analysis. (Likert, 1932)

A non-probability response scale is chosen because of its convenience and its limited financial requirements. However, some authors argue that this approach should only be used with caution because it is almost impossible to calculate the probability of getting the sample. Consequently, there is a need for application of some sociological knowledge when conducting the analysis. For this particular research, the respondents will be selected using ad-hoc-quotas non probability sampling. The latter approach will require that fifty percent of the respondents are male and fifty percent are female. Therefore as long as this quota is met, any respondent can be interviewed. (Likert, 1932)


The independent variables are those ones that cannot be manipulated and are simple measured. In this case, gender, cannot be manipulated and will be considered as an independent variables. Alternatively, prior victimization cannot be altered and also qualifies as an independent variable. The research will only measure these parameters by requesting respondents to indicate whether or not they have been victims of crime. On the other hand, fear of crime is affected by a number of factors as seen earlier. Consequently, this will be the dependent variable. (Babbie, 2005)

Gender is a nominal variable (one that cannot be quantified) so there will be no ranking in this case with reference to gender. This variable will only be used to classify answers i.e. qualitative analysis. On the other hand, prior victimization is an ordinal variable (one that can be ranked but cannot be allocated specific numbers or percentages). Consequently, questions in the structured interview will include rank prior victimization into categories i.e. those who have been victimized many times, those who have been victimized once or those how have never been victimized. (Meyers et al, 2005)

The ultimate goal in research is to establish a relationship between variables. In this particular research, fear of crime will be compare between men and women. If is found that men have higher fear of crime than women, then it ill be established that there is relationship between gender and ‘fear of crime’. (Meyers et al, 2005)


There are two key features that need to be considered when analyzing any sort of data; magnitude and reliability. Magnitude is established by determining the extent of relationships between two variables. In this case, the magnitude will be analyzed by determining how much gender affects fear of crime and also how much prior victimization affects fear of rime. If the research finds that there are higher numbers of women with high fear of crime, then it can be deduced that the magnitude of the two variables (fear of crime and gender) is high.

On the hand, reliability of the data cannot be analyzed directly through statistical methods. It employs the use of secondary data and prior knowledge in order to infer findings to the rest of the population. (Babbie, 2005)

After covering the definitions, the next challenge is on measurement of that magnitude. The number of variables covered in this research are three; gender, fear of crime and prior victimization. These variables will invariably determine method of analysis. Additionally, the Likert scale will also affect the type of analysis used and also the kind of relationships between these variables. (Babbie, 2005)

But the main principle behind measurement of these variables is a comparison between observable relations and the ideal relationship. This means that the analysis must involve a comparison between the extent of interrelation between the two parameters (gender and fear of crime) prior victimization and fear of crime) as indicated by the research and what should be common is the variables were in an ideal situation. (Meyers et al, 2005)

Taking an example of the effect of gender on fear of crime; if the proposed research reveals that the percentage of women who register high levels of fear of crime is sixty percent and the number of men to represent the percentage of men who register high fear of crime is forty, then we can analyze this by getting the deviation from the grand mean. In this case, the grand mean is fifty and the amount of deviation is ten.  In isolation, this number cannot indicate the relativity of the measurement. Consequently, it is important to lay out a reference point for the analysis. If set one hundred percent as the ideal value or the grand mean, then the analysis would yield totally different results. Thereafter, a deviation will need to be conducted relating the actual values of deviation in comparison to one hundred percent. The deviations found in the latter case would be extremely low because the grand mean chosen is too high. (Babbie, 2005)

Conversely, if the ideal was set as ten percent, the deviation between males and females in relation to fear of crime would be almost negligible for both genders and the research would find very low relations between gender and fear of crime. All in all, it is important to set the grand mean in relation to the kind of individual scores portrayed in the results in order to perform na accurate analysis. (Meyers et al, 2005)

However, before the calculation of deviation for the grand mean, there is a need to get a representative value form all the respondents’ answers. Since the data was collected using the Likert scale, then the first part of the analysis will also be done using methods limited to this scale. (Likert, 1932)

For this particular research, a number of items will be asked in relation to one particular issue. For instance, the responses asking people about their fear of crime in relation to specific crime types will be summed up and all the responses made by men will be analyzed. Similarly, women’s responses will be analyzed too. The summing up will assist in giving a summative score for the two types of genders. Thereafter, the different Likert items will be arranged into bar charts. The beauty about using bar charts is that they facilitate the calculation of a c3entral tendency. The proposed research will utilize median calculations because mean calculations will be use in later stages as seen above. Dispersion of the data will also be checked by percentage quartiles. (Likert, 1932)

It is permissible to sum up all the responses linking gender to crime because thy will all be utilizing the same scale. This means that it is an interval scale and much can be achieved though this. The same analysis will also need to be applied to prior victimization. All items relating to this parameter will be summed up and their correlation values determined through standard deviation. Thereafter, there will be a comparison of the two parameters; gender and prior victimization to fear of crime. The correlation values for gender and fear of crime will be compared to correlation values for prior victimization and fear of crime. Once it has been established that the values are somewhat similar, and then one can conclude that the two factors both affect fear of crime. But if one correlation value is much higher than the other (say gender is higher than prior victimization) then one can deduce that gender is the principle factor or vice versa if results favor prior victimization. (Likert, 1932)


The research will have a project tem made up of five people. The project coordinator will be responsible for coordinating all the activities necessary in the research. First of all, the team will start with plan. This will be a guide for all the activities that will take place within the time span of the project. Thereafter, material necessary for the research will be collected. As it had been stated earlier, there will be numerous research methods used and secondary sources re included. Library material will be identified and studied. After collection of all secondary material, then the primary research will begin. This will probably be done after a period of four weeks. (Meyers et al, 2005)

The project coordinator will be responsible for ensuring that al the interviewers are well aware of what they should do one they go out in the field. There will be no strict criteria for determining the kind of respondents as they ill employing non probability sampling. However, they must make sure that they cover equal numbers of men and women. (Meyers et al, 2005)

Before the interviewers go out in the field, they will sit down and agree on the questions. They will also ensure that there is an even balance between the negative and positive statements in order to minimize biased responses from the interviewees. This will last for two months

After the primary research is completed, the statistical work will begin and this may last for two weeks. Thereafter a conclusion will be done.


Research budget


Research coordinator’s logistics-$ 200

Interviewer’s logistics-$ 500

Project secretary-$ 100

Total- $800

Supplies and related expenses

Telephone calls-$ 300

Duplicating questions-$50

Computer-$ 1200

Office equipment-$420

Reference materials-$ 2000

Meeting facilities (temporary office)-$ 1000

Total-$ 4, 970


Staff travels-$ 1000

Total Expenses- $ 6770

Budget Narrative

The research will employ the services of five project members with one of them acting as a project coordinator. This coordinator will need to prepare meting venues, search for resources and at the same time meet his daily needs such as food. Because of these additional responsibilities, he will require higher payments hence the 200 dollars. The interviewers will each receive 125 dollars because thy have les responsibilities. This will be for their personal expenses such a food and accommodation. (Meyers et al, 2005)

The research process will require lots of secondary resources. (Babbie, 2005) Some of the journal items and books will purchased hence the two hundred dollars. On the other hand, internet resources will also be necessary hence the need for a computer. The project team will need a standard place to assemble because the preparation of research questions, data analysis, and data collection from primary resources will be quite time consuming. This is the reason why they will have to rent an office and they will also need office equipment for the stated amount. Lastly, project team members will need to travel to libraries, book stores and also to the respondents. Given the current transport economy it is likely the total will reach one thousand dollars. In conclusion, the proposed research project will require a total of six hundred, seven hundred and seventy dollars for completion.

10.0          REFERENCE:

Skogan, w. (1986): Fear of crime and neighborhood change; Crime and Justice, 8/, 203-229.

Hale, C. (1996): Fear of crime: A review of the literature; International Review of Victimology, 4, (2), 79-150

McGarrell,  F.,  Giacomazzi, A., & Thurman, Q. (1997): Neighborhood disorder, integration, and the fear of crime; Justice Quarterly, 14, (3), 479-500

Moore, S., & Shepherd, J. (2007): The elements and

prevalence of fear; The British Journal of Criminology,  47, (1), 154-162

Morris, A., Reilly, J., Berry, S., & Ransom, R. (2003): New Zealand National Survey of Crime Victims 2001; wellington: Ministry of Justice.

Acierno, R., Rheingold, A., Resnick, H., & Kilpatrick, G. (2004): Predictors of fear of crime in older adults; Journal of Anxiety Disorders,18, (3), 385-396

Akers, R. et al (1987): Fear of crime and victimization among the elderly in different types of communities; Criminology, 25, (3), 487-505.

Bennett, R., & Flavin, J. (1994): determinants of fear of crime: The effect of cultural setting; Justice Quarterly, 11, (3), 357-381

Ferraro, K. & LaGrange, I. (1988): Are older people afraid of crime? Journal of Aging Studies, 2, (3), 277-287

Stiles, B., Halim, S., & Kaplam, H. (2003): Fear of crime among individuals with physical limitations; Criminal Justice Review, 23, (2), 232-253

Warr, M. (1984): Fear of victimisation: why are women and the elderly more afraid?; Social Science Quarterly, 65,681-702

Statistics New Zealand (2006): Demographic trends 2005 (No.16.001). Wellington: Author

Australian Bureau of Statistics (ABS) (2006): Crime and safety, Australia, 2005 (No. 4509.0). Canberra: Author.

Babbie, Earl R. (2005): The Basics of Social Research; Thomson Wadsworth, p. 174

Likert, Rensis (1932): A Technique for the Measurement of Attitudes; Archives of Psychology 140: pp. 1-55

Meyers, L. et al (2005): Applied Multivariate Research: Design and Interpretation. Sage Publications Inc, p. 20